The report
How a trading bot fools you
Our agents looked good in testing. Then we tested them honestly, and most of them stopped trading. This is what we found in our own data, written so you can check any bot the same way: yours, ours or one somebody is selling you.
Free: the first chapter of our report
How a trading bot fools you. Read how our own test leaked, what we changed, and the ten red flags to check on any trading bot, before the full report opens. We email you the link, with one short desk summary a week after that. Unsubscribe in one click.
17 pages, about 6,000 words, with charts and tables, as a PDF.
Opens February 2027Leave your email and we'll tell you the day it opens. Nothing to pay now.
What is in it
- The leak. How days we called "unseen" had already been used to choose the winner, and how to tell if your own test has the same hole.
- Costs. What a trade really cost us against the few hundredths of a percent most tests assume.
- Stops. Why a stop order fills past its price, and what that does to a backtest that assumes it doesn't.
- Luck. Best of many tries looks like skill. How we now make an allowance for it.
- What happened when we fixed all four. The scores before and after, and how many agents were left trading.
- What survived: slower decisions and plain, long-published rules, with our ten-year test table.
- What the research says. Five published studies that found the same things on far more data.
- Ten red flags to look for when someone offers you a bot.
- Five tests you can run yourself, each a few lines of Python, on your own strategy.
- A one-page scorecard: ten questions for any trading bot, ours included.
Who it is for
Anyone building a trading bot, or thinking of paying for one. You don't need to code to follow it: the red flags and the scorecard are in plain words, and the five tests are there for whoever built the bot.
What it is not
It is not a strategy and it won't tell you what to buy. Every figure comes from MonsterDesk's own paper account and tests, with simulated money. Past results don't predict future ones.
The short version is free: read what our studies found, or watch the desk on the public record.